Most of the conventional array signal processing technologies are based on the assumption that targets are stationary. When the angle of arrival of targets actually moving is estimated by the technologies, performance deterioration occurs, and in particular, desired information on targets moving at a high speed may not be estimated due to a dispersion effect of the angle of arrival for an observation time.
To solve the above problem, as a method for positioning multiple targets moving at a high speed, a local polynomial approximation beamforming (LPA beamforming) method has been introduced. The LPA beamforming method is a method for simultaneously estimating an angle of arrival and an angular velocity by approximating a motion model of a target using an angle of arrival and an angular velocity and performing two-dimensional finding with the same resolution as a conventional beamformer (traditional beamforming method).
The method may also divide targets moving at different angular velocities at the same angle of arrival by estimating both of the angle of arrival and the angular velocity of the multi-target. However, the LPA beamforming method has the same resolution as the conventional beamformer and therefore may hardly differentiate closely moving targets.